ΔNp63α Expression Is Regulated by the Phosphoinositide 3-Kinase Pathway*

p63 is a homologue of p53 that functions to maintain progenitor cell populations in stratified epithelia. ΔNp63α is overexpressed in epithelial cancers and has been shown to have oncogenic properties. We have previously reported that inhibition of epidermal growth factor receptor signaling results in a decrease in ΔNp63α expression. Here, we demonstrate ΔNp63α is a target of the phosphoinositide-3-kinase (PI3K) pathway downstream of the epidermal growth factor receptor. Treatment of keratinocytes with epidermal growth factor results in an increase in ΔNp63α expression at the mRNA level, which is abrogated by inhibition of PI3K but not mitogen-activated protein kinase signaling. Small interfering RNA-mediated knockdown of the p110β catalytic subunit of PI3K results in a decrease in ΔNp63α protein levels in keratinocytes. The results presented herein suggest that regulation of ΔNp63α expression by the PI3K pathway plays a critical role in the survival and proliferative capacity of squamous epithelia.

p63 is a homologue of p53 that functions to maintain progenitor cell populations in stratified epithelia. ⌬Np63␣ is overexpressed in epithelial cancers and has been shown to have oncogenic properties. We have previously reported that inhibition of epidermal growth factor receptor signaling results in a decrease in ⌬Np63␣ expression. Here, we demonstrate ⌬Np63␣ is a target of the phosphoinositide-3-kinase (PI3K) pathway downstream of the epidermal growth factor receptor. Treatment of keratinocytes with epidermal growth factor results in an increase in ⌬Np63␣ expression at the mRNA level, which is abrogated by inhibition of PI3K but not mitogen-activated protein kinase signaling. Small interfering RNA-mediated knockdown of the p110␤ catalytic subunit of PI3K results in a decrease in ⌬Np63␣ protein levels in keratinocytes. The results presented herein suggest that regulation of ⌬Np63␣ expression by the PI3K pathway plays a critical role in the survival and proliferative capacity of squamous epithelia.
p63 is a recently discovered homologue of the tumor suppressor p53 (1,2). p63 is necessary to maintain progenitor cell populations in stratified epithelia. At least six different p63 transcripts can be expressed due to alternate promoter usage and alternative splicing (1,3); however, in epithelial cells, ⌬Np63␣ is the predominant form expressed (4 -8). The ⌬Np63␣ protein immunolocalizes to the proliferative, basal compartment of epithelia, including epidermis, oral mucosa, cervix, vaginal epithelium, urothelium, prostate, and breast (1,5,6). Expression of ⌬Np63␣ decreases in differentiating cells in vitro and in vivo, and studies suggest that ⌬Np63␣ is specifically expressed in epidermal stem cells possessing the highest proliferative capacity (4,9,10). Further insight to ⌬Np63␣ function is provided by animal models in which p63 expression is disrupted. Unlike p53 Ϫ/Ϫ mice, which are developmentally normal but rapidly develop tumors (11), p63 Ϫ/Ϫ mice display gross developmental abnormalities. The most striking of these is a complete lack of all stratified squamous epithelia and their derivatives, including epidermis, mammary glands, prostate, and other tissues (12,13). This phenotype is recapitulated in zebrafish, in which disruption of ⌬Np63 results in lack of epidermal morphogenesis (14,15). Taken together these data suggest that p63 is critical for the survival or proliferative capacity of epithelial stem cells.
⌬Np63␣ is overexpressed in several epithelial cancers often as a result of gene amplification (2, 16 -19). Overexpression of a ⌬Np63 isoform in Rat-1A cells increases colony growth in soft agar and xenograft tumor formation in nude mice, supporting the view that p63 acts as an oncogene (16). In addition, ⌬Np63␣ must be down-regulated for UV-B-induced apoptosis to occur (20). Several studies suggest that ⌬Np63␣ acts as a dominant negative with respect to p53 function, perhaps by competition for p53 DNA binding sites on downstream target genes (1,4). It is hypothesized that ⌬Np63␣ promotes the survival and maintenance of proliferative capacity of both epithelial stem cells and cancer cells.
The phosphoinositide 3-kinase (PI3K) 1 pathway is a well characterized positive regulator of cellular survival (21,22). PI3K is activated by receptor tyrosine kinases, such as the epidermal growth factor receptor (EGFR), and generates inositol phospholipids that act as second messengers to a variety of downstream targets (22). The PI3K pathway is deregulated in human cancers by a number of mechanisms, all of which promote survival of cancer cells. These include amplification of PI3K catalytic subunits, inactivation of the negative regulator PTEN, or overexpression of downstream targets such as Akt (21). Furthermore, the PI3K pathway is a key regulator of differentiation in a number of cell types, including keratinocytes (23).
We have previously reported that inhibition of EGFR signaling in head and neck cancer cells results in a decrease in ⌬Np63␣ expression (24). Here, we demonstrate that in primary and immortalized keratinocytes ⌬Np63␣ is a target of the PI3K pathway downstream of the EGFR. Treatment of keratinocytes with EGF results in an increase in ⌬Np63␣ expression, which is abrogated by inhibition of PI3K, but not MAPK signaling pathways. Furthermore, both decreased expression of the p110␤ catalytic subunit of PI3K by siRNA, and pharmacologic inhibition of PI3K result in a similar decrease in ⌬Np63␣ expression. Regulation of ⌬Np63␣ expression by the PI3K pathway may play a critical role in the survival and differentiation of squamous epithelia. human keratinocyte growth supplement S-001-5 (Cascade Biologics) and 0.06 mM CaCl 2 . The human keratinocyte cell line HaCaT was generously provided by P. Boukamp (German Cancer Research Center (DKFZ), Heidelberg, Germany). HaCaT cells were cultured in Dulbecco's modified Eagle's medium (DMEM, Invitrogen) supplemented with 10% fetal calf serum and 1% penicillin-streptomycin. All cells were cultured at 37°C with 5% CO 2 . For EGF treatment experiments, Ha-CaT cells were cultured in unsupplemented Dulbecco's modified Eagle's medium for 24 h and treated with 2 nM mouse epidermal growth factor (Invitrogen) for the indicated times. PI3K inhibitor LY294002 (Calbiochem) and MEK (mitogen-activated protein kinase/extracellular signalregulated kinase kinase) inhibitor PD98059 (Calbiochem) were used at 40 and 50 M final concentrations, respectively.
Flow Cytometry-Cells were trypsinized, and 1 ϫ 10 6 cells were analyzed by flow cytometry. The remaining cells were processed for protein analysis (see below). Cells were incubated with 50 g/ml propidium iodide (Sigma), and DNA content was measured using a FAC-SCaliber (BD Biosciences). Data were analyzed using Cell Quest software (BD Biosciences); 15,000 events were analyzed for each sample.

Treatment of Keratinocytes with EGF Increases Expression of
⌬Np63␣-We have previously reported that inhibition of EGFR signaling decreases expression of ⌬Np63␣ (24). To determine whether activation of the EGFR and its downstream signaling pathways would likewise increase expression of ⌬Np63␣, we treated the immortalized human keratinocyte cell line HaCaT with EGF. HaCaT cells were cultured without serum for 24 h. Cells were then treated with 2 nM EGF in serum-free media or left untreated for the indicated times. Growth factor deprivation resulted in a G 1 phase arrest with less than 5% of cells in S phase (Fig. 1A); treatment with EGF resulted in re-entry of cells into the cell cycle 12-18 h after treatment, with 30 and 37% of cells in S phase at 18 and 24 h, respectively (Fig. 1A). Western analysis showed an early increase in phosphorylated, active MAPK (Fig. 1B) after 5 min of EGF treatment, confirming activation of downstream signaling pathways. ⌬Np63␣ protein expression increased 6 h after EGF treatment (Fig. 1B) and preceded entry of the cells into S phase. Quantification of ⌬Np63␣ protein levels established that ⌬Np63␣ increased 2-3fold by 24 h after EGF treatment (Fig. 1, C and D). Northern analysis of p63 mRNA levels demonstrated that the increase in ⌬Np63␣ expression after EGF treatment occurred at the mRNA level (Fig. 1, E and F).
Inhibition of PI3K, but Not MAPK, Abrogates EGF-induced Increases in ⌬Np63␣ Expression-The PI3K and MAPK pathways are two of the most well characterized downstream effectors of EGFR signaling (28). To determine whether either of these signaling pathways is involved in EGFR regulation of ⌬Np63␣, we cultured HaCaT cells for 24 h in the absence of serum, then treated the cells with EGF alone or in combination with PD98059, an inhibitor of MAPK signaling, or LY294002, an inhibitor of PI3K. Cells were treated with Me 2 SO, PD98059 (50 M), or LY294002 (40 M) 30 min before treatment with EGF. Treatment with PD98059 eliminated EGF-induced MAPK activation ( Fig. 2A) and significantly decreased EGF-induced cell proliferation (Fig. 2B) but did not affect Akt phosphorylation (Fig. 2C). Inhibition of PI3K by LY294002 inhibited EGF-induced cell proliferation (Fig. 2B), and EGF-induced Akt phosphorylation ( Fig. 2C) but had no effect on MAPK activation ( Fig. 2A). Inhibition of PI3K eliminated EGF-induced increases in ⌬Np63␣ expression at both the protein ( Fig. 2A) and transcript levels (Fig. 2, D and E). However, inhibition of MAPK signaling by PD98059 had no effect on EGF-induced changes in p63 protein levels ( Fig. 2A). These results suggest that the PI3K pathway is required for EGF-induced increases in ⌬Np63␣ expression under the conditions assayed.
siRNA-mediated Knockdown of p110␤ Decreases ⌬Np63␣ Expression-To confirm that the PI3K pathway is required for regulation of ⌬Np63␣ expression, we utilized siRNAs to disrupt the expression of one of the catalytic subunits of PI3K, p110␤. This approach has previously been shown to decrease activation of downstream targets of the PI3K pathway (29). HaCaT cells were transfected with a hygromycin-selectable mammalian expression vector (pCEP-H1) encoding p110␤-targeting siRNAs under the control of the H1 promoter (26). Cells were selected for 7 days with hygromycin, and p110␤ and p63 protein levels were analyzed. Expression of two separate targeting sequences (p110␤-1 and p110␤-2) resulted in decreased expression of p110␤ that was accompanied by a significant decrease in ⌬Np63␣ expression (Fig. 3). Despite previous reports that siRNA-mediated knockdown of p110␤ reduced Akt activation (29), no effect was seen on phospho-Akt levels (data not shown). This is not surprising since Akt is a key regulator of cell survival and proliferation (21), and long term de-activation of Akt would be extremely detrimental to cells. We suspect that compensatory mechanisms (such as one of the other isoforms of p110) maintained Akt phosphorylation during the selection process in this experiment. These results confirm that specific ⌬Np63␣ Is Regulated by PI3K disruption of PI3K signaling leads to a reduction in ⌬Np63␣ expression.
⌬Np63␣ Expression Is Regulated by PI3K in Primary Keratinocytes-To confirm that ⌬Np63␣ was regulated by the PI3K pathway in cells with minimal genetic and epigenetic alterations, we utilized primary HEKs as a model system. HEKs were cultured for 24 h in basal media without growth factor supplements, then treated with 40 M LY294002 for the indicated times. Treatment with LY294002 resulted in a 70% decrease in ⌬Np63␣ expression over a 24-h period (Fig. 4, A and  B). It has been reported that down-regulation of ⌬Np63␣ in epidermal keratinocytes is required for apoptosis (20). Thus, we investigated if down-regulation of ⌬Np63␣ expression by inhibition of PI3K was associated with induction of apoptosis in HEKs. The timing and amount of ⌬Np63␣ down-regulation was accompanied by apoptosis in these cells as determined by the percentage of sub-diploid cells by flow cytometry (Fig. 4C, top) and poly(ADP-ribose) polymerase cleavage (Fig. 4C, bottom). These effects were only observed in the absence of growth factors. In the presence of growth factors, the decrease in ⌬Np63␣ expression and induction of apoptosis was minimal (data not shown). This implies that other signaling pathways are involved in maintenance of ⌬Np63␣ expression, but in conditions of cell stress (such as growth factor deprivation), ⌬Np63␣ expression is dependent on PI3K. These results demonstrate that ⌬Np63␣ is regulated by the PI3K pathway in HEKs, and reduction in ⌬Np63␣ expression is accompanied by induction of apoptosis in these cells.

DISCUSSION
⌬Np63␣ is a homologue of the tumor suppressor p53, which acts as a dominant negative with respect to p53 function and has an anti-apoptotic function in keratinocytes (1,4,20). Here, we report that ⌬Np63␣ expression is regulated by the PI3K pathway, a well characterized regulator of survival in a wide variety of cell types (21). We show that treatment of immortalized keratinocytes with EGF results in activation of both the MAPK and PI3K pathways and an increase in ⌬Np63␣ mRNA levels. The simplest explanation for the increase in ⌬Np63␣ expression is a change in p63 gene transcription; however, effects on message stability and post-transcriptional processing are also possibilities. Furthermore, we cannot rule out additional effects on protein stability or translational control. Pharmacologic inhibition of PI3K but not MAPK signaling abrogates this increase in ⌬Np63␣ expression. We have confirmed that this effect is specific for the PI3K pathway by disrupting PI3K signaling using siRNAs directed against a catalytic subunit of PI3K, p110␤, in immortalized keratinocytes. Finally, we have shown that ⌬Np63␣ is regulated by PI3K signaling in primary human keratinocytes and that decreased ⌬Np63␣ expression resulting from PI3K inhibition is accompanied by an induction of apoptosis in these cells.
The PI3K pathway is an important regulator of survival in a number of cell types, including keratinocytes. PI3K affects cell survival through a number of intermediates, the most extensively studied of which is Akt (PKB) (21). Akt is a serine/ threonine kinase that is activated by PI3K and inhibits cell death by directly phosphorylating several components of the apoptotic machinery (21). Although Akt is the most well characterized effector of PI3K-mediated survival, other downstream targets are also implicated, including cytokine-independent survival kinase, protein kinase C, and integrin-linked kinase (22,30). ⌬Np63␣ is implicated in the survival and maintenance of epithelial cells. In both mice and zebrafish, p63 is required for the maintenance of progenitor cells necessary for epidermal morphogenesis (12)(13)(14)(15). Furthermore, several studies show that ⌬Np63␣ can act as a dominant negative with respect to p53 function and can potentially repress p53 target genes (1,4,31). In addition, ⌬Np63␣ is down-regulated in response to UV-B irradiation, and targeted expression of ⌬Np63␣ in murine skin inhibits UV-B induced apoptosis (20). We observed that reduction in ⌬Np63␣ expression by inhibition of PI3K correlates with induction of apoptosis in primary epidermal keratinocytes; however, it is uncertain whether this represents an anti-apoptotic role for ⌬Np63␣ or if ⌬Np63␣ is simply a marker for the viability or proliferative capacity of cells. The precise role of ⌬Np63␣ in cellular survival and apoptosis is still unclear, but the discovery that ⌬Np63␣ is regulated by the PI3K pathway further implicates it in these processes.
Because of its potent anti-apoptotic effects, the PI3K pathway is often altered in human cancers (21). The deregulation of this pathway can occur at many levels. These include amplification of PI3K catalytic subunits (33,34), inactivation of negative regulators such as PTEN (35), or overexpression of downstream targets such a Akt (36,37). ⌬Np63␣ is overexpressed in several epithelial cancers, including carcinomas of the head and neck, lung, esophagus, bladder, and cervix (16 -19, 38). This overexpression can sometimes be the result of amplification of the p63 genomic locus (16). In fact, p110␣, one of the catalytic subunits of PI3K, and p63 are both overexpressed in squamous cell carcinomas of the head and neck (16,17,39,40). Further supporting the view that p63 may act as an oncogene, overexpression of a ⌬Np63 isoform in Rat-1A cells increases colony growth in soft agar and xenograft tumor formation in nude mice (16). These studies highlight similarities between the PI3K pathway and p63 in human cancers.
It is well established that ⌬Np63␣ expression decreases during the differentiation of epidermal keratinocytes in vitro and in vivo (4,9,10,12,41). One study suggests that ectopic expression of ⌬Np63␣ is sufficient to block keratinocyte differentiation (42), but the functional role of ⌬Np63␣ in this biological process remains unclear (20,41). The PI3K pathway is also implicated in the differentiation of a wide variety of cell types (43)(44)(45)(46). In epidermal keratinocytes, blockade of PI3K results in differentiation, and activation of PI3K prevents differentiation (23). In this study we report that blockade of PI3K results in decreased expression of ⌬Np63␣ (which occurs during differentiation) and that activation of PI3K results in increased expression of ⌬Np63␣ (which has been reported to block differentiation). These observations suggest that the regulation of ⌬Np63␣ expression by PI3K signaling may play a role in the differentiation of epidermal keratinocytes, but further investigation is necessary to establish this connection.
A wide variety of upstream stimuli activate the PI3K pathway, leading to the potential for a large number of extracellular signals to affect ⌬Np63␣ expression. PI3K is activated by a diverse collection of growth factors, including EGF, insulin, insulin-like growth factor-1, and neurotrophins (28,(47)(48)(49). We found that in primary keratinocytes the presence of growth factors limits the reduction in ⌬Np63␣ by PI3K inhibition. This implies that other signaling pathways are involved in maintenance of ⌬Np63␣ expression, but in conditions of cell stress (such as growth factor deprivation) ⌬Np63␣ expression is heavily dependent on PI3K. This is not surprising, since the PI3K pathway is important for the survival of cells in response to stresses.
Another key upstream activator of the PI3K pathway is adhesion molecule signaling. Integrins and other mediators of cell-matrix interaction activate PI3K through focal adhesion kinase, Ras, and other intermediates (50), and disruption of integrin-extracellular matrix contact eliminates PI3K activity (51,52). This regulation of PI3K by integrin signaling has particular implications for ⌬Np63␣ expression. ⌬Np63␣ expression in the epidermis is limited to the basal layer of keratinocytes, which maintain contact with the basement mem-brane (12,13); certain integrin signaling pathways are highly active in these cells (32,53). Integrin-mediated activation of the PI3K pathway, therefore, may be partly responsible for the observed distribution of ⌬Np63␣ expression in the epidermis. The large number of upstream signals that can activate PI3K suggest that although we used EGF to interrogate the PI3K pathway in vitro, many different factors may coordinately regulate ⌬Np63␣ expression through PI3K in vivo.
In summary, we show that ⌬Np63␣ is regulated by the PI3K pathway in primary and immortalized squamous epithelial cells. This discovery provides a framework for investigation of how biological processes such as growth factor and adhesion molecule signaling can affect expression of ⌬Np63␣. In addition, regulation of ⌬Np63␣ expression by the PI3K pathway may play a critical role in the survival and differentiation of squamous epithelia.